Abstract
Decades of research have demonstrated that the hippocampus is a central structure in the function of episodic memory. The phenomenology of the hippocampus has components that could conceivably represent the “what” and “where” components of the three-part what/when/where model of episodic memory, but there is relatively little evidence of hippocampal phenomenology that could represent the “when” component on timescales longer than a few minutes. The aim of the present thesis was to investigate how the hippocampus might represent temporal information over longer timescales. In order to investigate how this might happen, the firing rate of hippocampal CA1 place cells and hippocampal EEG in the theta-band were examined over very long recordings (25-30h). It was found that the firing rate of CA1 place cells oscillated on a circadian time period, and that this firing rate was not merely entrained to light as might be expected from a signal entrained to the master oscillator in the SCN, but instead appeared to be entrained with some variable coincident with the start of recording. It was also found that the frequency but not the power of hippocampal theta rhythm oscillated in an identical fashion. It was therefore hypothesised that some variable coincident with the start of the recording sessions served as an entraining stimulus. An experiment in which animals were switched into a novel environment halfway through a 25-hour recording demonstrated that environmental novelty was not sufficient to re-entrain the observed oscillation. It was therefore hypothesised that the availability of food may be the entraining stimulus. Using a novel paradigm that involved the driving of hippocampal theta by electrical stimulation of the reticular formation, the presentation of food was disambiguated from the other variables coincident with the start of recording. It was found that the frequency and power of reticular activated theta (R.A.T) oscillated on a 25-hour cycle, and that food was a sufficient zeitgeber for this oscillation. These results are considered in a model through which the newly observed circadian modulation in place cell activity might represent a temporal “time field” in the same way as it has been hypothesised that these cells may function in a spatial context, and the association of single unit firing rate and theta frequency might represent the hippocampus engaged in memory engram storage and/or retrieval. The observed oscillations in hippocampal activity reported in the present thesis might therefore represent a mechanism through which the missing “when” component of episodic memory is represented in the hippocampus.